Method and circuit for controlling setup of multichannel system

Abstract

A circuit and method that optimizes a set up time and control of bit and energy loadings of data sub-channels in a multi-channel data transmission signal is disclosed. A time out period, or an iteration count are used to control a setup time. An initialization routine attempts to make appropriate loadings within such time out period or within an iteration count parameter. If it is unable to do so, a handoff is made to a second routine to complete the loadings. In a preferred embodiment the initialization routine calculates rough parameters, while the second routine finalizes the capacity calculations. Another aspect of the invention ensures that the initialization routine can be repeated at a later time to re-compute appropriate bit and energy loadings to ensure that significant changes caused by line disturbances or large data rate variations can be quickly accommodated. In a preferred embodiment the resulting bit/energy loadings can be adjusted to be fully compliant with applicable Discrete Multi-Tone (DPM implementations of Asymmetric Digital Subscriber Loop (ADSL) protocols.

What is claimed is:  
     
       1. A circuit for use in a high speed multi-channel transmission system, which system is intended to transmit data at a rate R using K sub-channels through a channel having varying transmission characteristics, said circuit comprising: 
       a sub-channel parameter memory for storing data capacities associated with some or all of said sub-channels, including initial data capacities and adapted data capacities; and  
       a processing unit which can perform a configuration procedure to generate said initial data capacities to configure said system to operate at the data rate R; and  
       said processing unit being further configured to repeat said configuration procedure later during a data transmission to generate said adapted data capacities when significant changes are required in the data rate R based on an evaluation of one or more of the following parameters:  
       (a) signal-to-noise ratio for the K sub-channels;  
       (b) variations in said transmission characteristics;  
       (c) variations in said data rate; and  
       (d) a specified allowable system adaptation time period; and  
       wherein the processing unit executes the configuration procedure until said data rate is achieved or until a specified iteration criteria parameter is satisfied and thereafter repeatedly executes said configuration procedure as needed to maintain the data rate R.  
     
     
       2. The circuit of claim  1 , wherein said first adaptation routine is selected by said processing unit when variations in said signal-to noise ratios are relatively large. 
     
     
       3. The circuit of claim  2 , wherein said processing unit selects a separate adaptation routine when the signal-to-noise ratios and data rate R variations are relatively small. 
     
     
       4. The circuit of claim  3 , wherein said separate adaptation routine is selected by said processing unit when a specified allowable system adaptation time period is relatively small. 
     
     
       5. A method of controlling data transmissions through a channel, the method comprising the steps of: 
       (a) performing a bit and gain loading routine for initializing the data transmissions, said bit and gain loading routine including:  
       measuring initial signal to noise ratio information for sub-channel data carriers to be used for the data transmission;  
       storing said initial signal to noise information in an array; and  
       configuring initial capacity parameters for said sub-channel data carriers based on said initial signal to noise ratio information;  
       (b) monitoring the channel for changes in transmission characteristics;  
       (c) repeating said bit and gain loading routine as needed when there relatively large changes in said initial signal to noise information.  
     
     
       6. The method of claim  5 , wherein said bit and gain loading routine is executed for a number of predetermined iterations, and, if such routine is not able to complete configuring said initial capacity parameters, a second routine is used for attempting to set up a data transmission in the channel. 
     
     
       7. The method of claim  5 , wherein said bit and gain loading routine includes a non-iterative portion for achieving approximate bit and gain loading data, and an iterative portion for converging said bit and gain loading data. 
     
     
       8. The method of claim  5 , wherein a time required to perform said bit and gain loading routine during step (c) is not directly related to changes required in said initial capacity parameters. 
     
     
       9. A method of controlling data transmissions through a channel, the method comprising the steps of: 
       (a) performing an initialization routine to initialize the channel, which initialization routine includes a bit and gain loading routine that calculates and configures initial bit values and initial gain values for a number of sub-channels used for the data transmissions;  
       (b) monitoring the channel for changes in transmission characteristics that could affect said initial bit and initial gain values; and  
       (c) repeating said bit and gain loading routine in response to step (b) so as to calculate new bit and gain values for said number of sub-channels when there are significant changes in said transmission characteristics that affect said initial bit values and said initial gain values wherein said bit and gain loading routine is executed for a number of predetermined iterations, and, if such routine is not able to complete configuring said initial capacity parameters, a second routine is used for attempting to set up a data transmission in the channel.  
     
     
       10. The method of claim  9 , wherein said bit and gain loading routine includes a non-iterative portion for achieving approximate bit and gain loading data, and an iterative portion for converging said bit and gain loading data. 
     
     
       11. The method of claim  9 , wherein a time required to perform said bit and gain loading routine during step (c) is not directly related to loading changes required in said initial bit values.